Tuesday
There were two interesting themes in the solar sessions this morning. The first was a really positive story about how instrumental differences between rival (and highly competitive) teams can get resolved. This refers to the calibration of measurements of the Total Solar Irradiance (TSI). As is relatively well known, the different satellite instruments over the last 30 or so years have shown a good coherence of variability – especially the solar cycle, but have differed markedly on the absolute value of the TSI (see the figure). In particular, four currently flying instruments (SORCE, ACRIM3, VIRGO and PREMOS) had offsets as large as 5W/m2. However, the development of a test-facility at NASA Langley the
University of Colorado, Laboratory for Atmospheric and Space Physics in Boulder
Colorado – an effort led by Greg Kopp’s group – has allowed people to test their instruments in a vacuum, with light levels comparable to the solar irradiance, and have the results compared to really high precision measurements. This was a tremendous technical challenge, but as Kopp stated, getting everyone on board was perhaps a larger social challenge.
The facility has enabled the different instrument teams to calibrate their instruments, and check for uncorrected errors, like excessive scattering and diffusive light contamination in the measurement chambers. In doing so, Richard Wilson of the ACRIM group reported that they found higher levels of scattering than they had anticipated, which was leading to slightly excessive readings. Combined with a full implementation of an annually varying temperature correction, their latest processed data product has reduced the discrepancy with the TIM instrument from over 5 W/m2 to less than 0.5 W/m2 – a huge improvement. The new PREMOS instrument onboard Picard, a french satellite, was also tested before launch last year, and they improved their calibration as well – and the data that they reported was also very close to the SORCE/TIM data: around 1361 W/m2 at solar minimum.
The errors uncovered and the uncertainties reduced as a function of this process was a great testament to the desire of everyone concerned to work towards finding the right answer – despite initial assumptions about who may have had the best design. The answer is that space borne instrumentation is hard to do, and thinking of everything that might go wrong is a real challenge.
The other theme was the discussion of the spectral irradiance changes – specifically by how much the UV changes over a solar cycle are larger in magnitude than the changes in the total irradiance. The SIM/SOLSTICE instruments on SORCE have reported much larger UV changes than previous estimates, and this has been widely questioned (see here for a previous discussion). The reason for the unease is that the UV instruments have a very large degradation of their signal over time, and the residual trends are quite sensitive to the large corrections that need to be made. Jerry Harder discussed those corrections and defended the SIM published data, while another speaker made clear how anomalous that data was. Meanwhile, some climate modellers are already using the SIM data to see whether that improves the model simulations of ozone and temperature responses in the stratosphere. However, the ‘observed’ data on this is itself somewhat uncertain – for instance, comparing the SAGE results (reported in Gray et al, 2011) with the SABER results (Merkel et al, 2011), shows a big difference in how large the ozone response is. So this remains a bit of a stumper.
The afternoon sessions on water isotopes in precipitation was quite exciting because of the number of people looking at innovative proxy archives, including cave records of 18O in calcite, or deuterium in leaf waxes, which are extending the coverage (in time and space) of this variable. Even more notable, was the number of these presentations that combined their data work with interpretations driven by GCM models that include isotope tracers that allow for more nuanced conclusions. This is an approach that was pioneered decades ago, but has taken a while to really get used routinely.
References
- L.J. Gray, J. Beer, M. Geller, J.D. Haigh, M. Lockwood, K. Matthes, U. Cubasch, D. Fleitmann, G. Harrison, L. Hood, J. Luterbacher, G.A. Meehl, D. Shindell, B. van Geel, and W. White, "SOLAR INFLUENCES ON CLIMATE", Reviews of Geophysics, vol. 48, 2010. http://dx.doi.org/10.1029/2009RG000282
- A.W. Merkel, J.W. Harder, D.R. Marsh, A.K. Smith, J.M. Fontenla, and T.N. Woods, "The impact of solar spectral irradiance variability on middle atmospheric ozone", Geophysical Research Letters, vol. 38, pp. n/a-n/a, 2011. http://dx.doi.org/10.1029/2011GL047561
One of the best prospects to understand the complete solar – Earth link is a recent discovery (m.a.vukcevic) of the direct but non-stationary relationship between the heliospheric magnetic field at the Earth’s orbit and the CET’s ‘reference’ component:
http://www.vukcevic.talktalk.net/HMF-T.htm
The revision of the ACRIM data sounds very interesting, especially in the context that the ACRIM composite has been used by some (esp. Dr. Scafetta) to suggest that a jump in the apparent solar output during the so-called “ACRIM gap” of 1989 was an indication that solar output had risen somewhat through the “90’s” and thus an increase in solar output could be “rescued” as a major contribution to global warmng in that period!
[Response: The ACRIM gap is still an issue, though I think there is more support for the PMOD method than the official ACRIM version. Scafetta’s attempt to deal with it, was comprehensively demolished by Krivova et al. – gavin]
So does the recent Wilson revision of ACRIM resolve that particular issue or is there still uncertainty in the progression of solar output through the ACRIM gap?
This particular issue is discussed very informatively here:
http://rspa.royalsocietypublishing.org/content/464/2094/1367.abstract
This poster would fit in well with some recent discussions on the unforced variations threads for Dec.
http://eposters.agu.org/files/2011/12/Hausfather-et-al-Urban-Trends.pdf
Do not meddle in the affairs of photometrists, for they are subtle and quick to anger.
In other AGU news, Jules and James have the culinary report and this comment regarding the modeling philosophy session:
Steve Easterbrook is solid on models.
Hausfather, et. al. have a poster appropriate to our recent discussion on the unforced variations thread.
http://eposters.agu.org/abstracts/the-impact-of-urbanization-on-global-surface-temperature-trends/?from_search=true
They improved their flux calibration from 0.36% to 0.036%???? That’s majorly impressive.
So — for us non-experts, is the level of solar irradiance now accepted to be about 5 W/m2 below the pre-2011 satellite estimates? I’m not trying to grind an ax of any kind, just trying to update my current understanding.
[Response: Yes. The TSI at solar minimum should be updated to around 1361 W/m2 (uncertainty around 0.5 W/m2). – gavin]
And for yet other even less expert headscratchers, what is the most recent trend in irradiance–is it sharply up in the last few months? Is this measure independent of sunspot activity?
Dan H.,
My poster was on day 3 of the AGU, rather than day 2. Also, while our number is higher than some past estimates, we only show a 10% difference between urban and rural stations over the past three decades, which translates into closer to a 5% difference between rural and all stations. Add in the fact that over 2/3rds of the earth is ocean, and you only have a global temp bias due to UHI of 1-2%.
We also only examined raw GHCN-Daily data. The NCDC folks are in the process of homogenizing it for us (previously only GHCN-Monthly has been homogenized), and we expect that at least some of the urban bias can be detected by the pair-wise homogenization process and removed (as we saw in the US).
So as we would expect there seems to be a positive bias due to urbanization, but the best that we can tell it is relatively modest.
Zeke,
Thank you for your reply. There wasn’t a thread for day 3 yet, so I posted the link here. The ocean issue is a given, but most temperature data show higher warming over land than ocean.
The argument began when a poster claimed that there was no bias. There has been disagreement as to whether rural or urban stations are warming faster. The claim by a few of us is that the bias occurs when a station shifts from being rural to urban.
> the claim by a few of us
Muller, e.g. in his testimony to Congress, tested that:
“bias … for example, a result of choosing long continuous records. (A long record might mean a station that was once on the outskirts and is now within a city.)
To avoid such station selection bias … work with all the available stations ….
…
… will largely eliminate the station selection bias …”
berkeleyearth.org/Resources/Muller_Testimony_31_March_2011
As he says, without any adjustments, his results are consistent with those previously published by others.
Dan H.
Interestingly enough, the BEST folks revised their estimate of UHI to 0.02 C per decade from 1950-present. Its not statistically significantly different from zero in their estimate, but its not negative anymore.
Zeke Hausfather #12, have you got a link for the revised BEST UHI estimate, please?a
@gavin: If the TSI is updated from 1365W/m² to 1361W/m² how does this affect the estimates for reflected and absorbed solar energy? Does only the reflected energy going down? Or the absorbed solar energy? Or may the albedo estimates remain constant so affecting both? Or is it more difficult?
[Response: The reflected solar energy is a directly measured quantity, and so a change in the incoming TSI will impact the calculated albedo. But the differences are not large. For the 5 W/m2 change in TSI, that makes 1.25 W/m2 change in net incoming radiation (1.25/342 or 0.4%), and that makes 0.02% difference in the derived albedo (i.e. from 29.8 to 30%) (assuming 240 W/m2 is absorbed). – gavin]
@gavin: Thanks! But could it be that the same scattering problem has also affected the measurements of the reflected solar energy? If yes, then the value for the reflected solar energy would have to reduced by the same percentage and the albedo does not change. Or could this be excluded, maybe because of completely different instrument types?